Perl 6 language and compiler development | Logs at colabti.org/irclogger/irclogger_log/perl6-dev | For toolchain/installation stuff see #perl6-toolchain | For MoarVM see #moarvm
Set by Zoffix on 27 July 2018.
00:01 MasterDuke left
pmurias timotimo: re non serializable variable, I don't think there is a good way to do that withough adding special magic to the sixmodel in the backends 00:11
timotimo: if I remember correctly I think state variables don't get serialized so you could abuse that 00:12
timotimo nah, that won't work easily for what i want
but yeah, the compose protocol for p6opaque could get a key for "attribute name that isn't supposed to participate in serialization" 00:13
is probably enough for my purpose to have only a single one, even
wakelift.de/2018/11/10/rakudo-perl...ss-report/ - lol i blogged a second time! 00:23
00:26 MasterDuke joined, p6bannerbot sets mode: +v MasterDuke, MasterDuke left, MasterDuke joined, herbert.freenode.net sets mode: +v MasterDuke, p6bannerbot sets mode: +v MasterDuke 00:28 j3nnn1 left 00:40 pmurias left 00:50 SyrupThinker left 01:35 MasterDuke left 01:57 MasterDuke joined, p6bannerbot sets mode: +v MasterDuke 02:04 MasterDuke left, MasterDuke joined, herbert.freenode.net sets mode: +v MasterDuke, p6bannerbot sets mode: +v MasterDuke 02:14 delon joined, p6bannerbot sets mode: +v delon 03:02 entonian joined 03:03 p6bannerbot sets mode: +v entonian 03:04 entonian left 05:38 delon left
lizmat Files=1255, Tests=76374, 345 wallclock secs (15.97 usr 5.40 sys + 2437.26 cusr 226.39 csys = 2685.02 CPU) 07:57
08:42 AlexDaniel left 10:27 AlexDaniel joined, p6bannerbot sets mode: +v AlexDaniel 10:43 pmurias joined, p6bannerbot sets mode: +v pmurias 10:59 pmurias left 11:00 pmurias joined, p6bannerbot sets mode: +v pmurias
Geth rakudo: c50cf82afc | (Elizabeth Mattijsen)++ | src/core/Hyper.pm6
An initial stab at handling <<op=>> faster
 13:21
13:34 AlexDaniel left
b2gills timotimo: It was difficult trying to figure out which color was supposed to be associated with which verb (promoted,kept,freed) in the combined graphs. Perhaps adding a colored "\c[full block]" would help. 14:05
14:44 AlexDaniel joined, p6bannerbot sets mode: +v AlexDaniel 14:48 MasterDuke left
SqrtNegInf m: my @A = [[1], [2, 3]]; my @B = @A »*» 0; @B[0][0] = 42; 15:13
camelia Cannot assign to an immutable value
in block <unit> at <tmp> line 1
SqrtNegInf m: my @A = [[1], [2, 3]]; my @B = @A; @B[0][0] = 42; 15:14
camelia ( no output )
AlexDaniel m: sqrt -∞ 15:19
camelia WARNINGS for <tmp>:
Useless use of "sqrt -∞" in expression "sqrt -∞" in sink context (line 1)
AlexDaniel m: say sqrt -∞
camelia NaN
SqrtNegInf m: say (-∞+0i)**½ 15:22
camelia Inf+Inf\i
SqrtNegInf That aside, a recently created bug with hypers, open an issue about it? 15:23
lizmat SqrtNegInf: please do 15:32
I'm probably to blame 15:33
bisectable6: my @A = [[1], [2, 3]]; my @b = @A »*» 0; @b[0][0] = 42; 15:36
bisectable6 lizmat, Bisecting by exit code (old=2015.12 new=c50cf82). Old exit code: 0
lizmat, bisect log: gist.github.com/34f90a4531e87fac23...b9b3a4370f 15:37
lizmat, There are 6 candidates for the first “new” revision. See the log for more details
lizmat yeah, definitely my fault 15:38
SqrtNegInf Opened issue, github.com/rakudo/rakudo/issues/2480 15:40
lizmat SqrtNegInf: working on it 15:48
Geth rakudo: lizmat self-assigned Array built with hyper operations becomes immutable github.com/rakudo/rakudo/issues/2480
1a2b39dc1a | (Elizabeth Mattijsen)++ | src/core/Hyper.pm6

Arrays were created, but without containers. Fixes R#2840. This **also** fixes the case where the original array was typed: the type *and* its associated descriptor are cloned.		 15:49
lizmat bisectable6: my Int @A = [1,2,3]; my @b := @A >>+>> 0; @b[0] = "foo"; say @b 16:07
bisectable6 lizmat, Bisecting by output (old=2015.12 new=c50cf82) because on both starting points the exit code is 1
lizmat I guess that error has been there all along...
bisectable6 lizmat, bisect log: gist.github.com/71908c67bb9e9b4f5a...faa4980aa3
lizmat, (2016-02-18) github.com/rakudo/rakudo/commit/99...4283b48d0a
lizmat that's a weird bisect 16:08
AlexDaniel c: 9983c2c8^,9983c2c8 my Int @A = [1,2,3]; my @b := @A >>+>> 0; @b[0] = "foo"; say @b
committable6 AlexDaniel, gist.github.com/fecf57dfb541c071ca...7361fb2cb5
AlexDaniel c: old=2018.10 my Int @A = [1,2,3]; my @b := @A >>+>> 0; @b[0] = "foo"; say @b 16:09
committable6 AlexDaniel, ENV variable old is not supported
AlexDaniel bisect: old=2018.10 my Int @A = [1,2,3]; my @b := @A >>+>> 0; @b[0] = "foo"; say @b
bisectable6 AlexDaniel, Bisecting by output (old=2018.10 new=c50cf82) because on both starting points the exit code is 1
AlexDaniel, bisect log: gist.github.com/3597e375b042d737da...cf4c4c62cc
AlexDaniel, There are 6 candidates for the first “new” revision. See the log for more details
benjikun What's weird about this? 16:10
the immutable value one? 16:11
lizmat yeah, that commit doesn't even come near any meta ops
benjikun hmmm
AlexDaniel generally it will find the first change
“Bisecting by output” meaning any change to the error message is “new” 16:12
you can see here what changed exactly: gist.github.com/fecf57dfb541c071ca...7361fb2cb5
so the bisect result is correct, but yeah, that is becoming a problem somewhat. 2015.12 is too old to bisect blindly 16:13
lizmat yeah... I guess
AlexDaniel I'm hoping to reset it to 2018.11 if that has v6.d
by default I mean
I mean, change it
SqrtNegInf That bug showed up just yesterday in a program that's run without errors for over two years (in daily smoke-testing)
lizmat anyways, I think I found a solution for the original R#2480
synopsebot R#2480 [open]: github.com/rakudo/rakudo/issues/2480 [⚠ blocker ⚠] Array built with hyper operations becomes immutable
lizmat SqrtNegInf: well, there was no spectest for it, so creating a roast PR would be greatly appreciated 16:14
SqrtNegInf Will do 16:15
lizmat SqrtNegInf++
AlexDaniel fwiw to bisect without surprises: first run `6c: code here` to see what is going on across releases, then bisect with `bisect: old=…,new=… code` with starting points of your interest, then double check with `c: commit^,commit code` 16:16
arguably the last step should be done automatically… 16:17
benjikun SqrtNegInf: that chemical structure regular expressions thing you made is cool 16:18
SqrtNegInf Upside: done for $work, so paid for doing interesting stuff. Downside: done in Fortran, because Chemistry. 16:23
lizmat SqrtNegInf: ^^^ if you could also add tests for: 16:25
m: my Int @a = ^10; my @b := @a <<+>> 0; @b[0] = "foo"; dd @b
camelia Cannot assign to an immutable value
in block <unit> at <tmp> line 1
benjikun your job must be cool
SqrtNegInf lizmat: Got it 16:29
benjikun: Among nerdly types, perhaps. But it also gives me a super-power, making Eyes Glaze Over in milliseconds. 16:31
benjikun well, your work certainly seems respectable from my eyes :) 16:33
Geth rakudo: 307a98bdb6 | (Elizabeth Mattijsen)++ | src/core/Hyper.pm6
We should always .STORE unless it's a List itself

Simplify the check for that		 16:51
lizmat AlexDaniel: should or shouldn't I remove the blocker label on R#2840 ? 16:52
AlexDaniel lizmat: if you tested it manually and you think that it is fixed, then yes, you can remove the label 16:54
lizmat AlexDaniel: done
AlexDaniel lizmat++
17:01 MasterDuke joined, p6bannerbot sets mode: +v MasterDuke, MasterDuke left, MasterDuke joined, herbert.freenode.net sets mode: +v MasterDuke, p6bannerbot sets mode: +v MasterDuke
japhb SqrtNegInf: What is the chemistry thing that benjikun is referring to? 17:22
17:26 travis-ci joined, p6bannerbot sets mode: +v travis-ci
travis-ci Rakudo build failed. Elizabeth Mattijsen 'Make sure we containerize arrays in @a <<+>> 0 17:26
travis-ci.org/rakudo/rakudo/builds/453308574 github.com/rakudo/rakudo/compare/c...2b39dc1a9b
17:26 travis-ci left
buggable [travis build above] ☠ Did not recognize some failures. Check results manually. 17:26
benjikun japhb: github.com/SqrtNegInf/SMARTS here it is 17:40
Geth rakudo: jmaslak++ created pull request #2481:
Remove redundant smartmatch usage in IO::Socket::Async		 17:41
18:01 ExtraCrispy left
[Tux] Rakudo version 2018.10-103-gc50cf82af - MoarVM version 2018.10-74-g2fdde4a21
csv-ip5xs0.930 - 0.947
csv-ip5xs-207.326 - 7.358
csv-parser24.746 - 25.457
csv-test-xs-200.429 - 0.439
test7.729 - 8.733
test-t1.895 - 1.956
test-t --race0.855 - 0.933
test-t-2033.086 - 33.514
test-t-20 --race12.379 - 13.564
 18:31
20:06 delon joined 20:07 p6bannerbot sets mode: +v delon
dogbert11 m: say 10 <<*<< (1 .. 4) 20:17
camelia Cannot modify an immutable Range (1..4)
in block <unit> at <tmp> line 1
timotimo yeah, that's not right 20:19
good catch
dogbert11 I ran spectest :)
timotimo d'oh 20:20
dogbert11 was spectesting the stuff nine++ just fixed 20:22
Geth rakudo: 5522ee90a0 | (Joelle Maslak)++ | src/core/IO/Socket/Async.pm6
Remove redundant smartmatch usage in IO::Socket::Async
 21:35
rakudo: 76d2fde4ba | (Jonathan Worthington)++ (committed using GitHub Web editor) | src/core/IO/Socket/Async.pm6
Merge pull request #2481 from jmaslak/Async-Port-Constraint-Optimize

Remove redundant smartmatch usage in IO::Socket::Async
rakudo: 843118ea34 | (Elizabeth Mattijsen)++ | src/core/Hyper.pm6
Revert "We should always .STORE unless it's a List itself"

This reverts commit 307a98bdb6d824c893127bdb8d64b4b9577f8835.
Turns out Ranges are also Positional. So until we have a candidate that handles Ranges, we should revert to the original condition. dogbert++ for spotting.		 21:54
lizmat dogbert11 timotimo ^^^
afk again& 21:55
dogbert11 lizmat++ 22:25
22:58 pmurias left 23:34 travis-ci joined, p6bannerbot sets mode: +v travis-ci
travis-ci Rakudo build failed. Jonathan Worthington 'Merge pull request #2481 from jmaslak/Async-Port-Constraint-Optimize 23:34
travis-ci.org/rakudo/rakudo/builds/453388148 github.com/rakudo/rakudo/compare/3...d2fde4ba53
23:34 travis-ci left
buggable [travis build above] ☠ Did not recognize some failures. Check results manually. 23:34

Powered by Cro and the Raku® Programming Language.

Please report any issues / comments / feature requests as an issue on App::Raku::Log.

Thank you!